Skip to main content

Home/ Advanced Concepts Team/ Group items tagged neural

Rss Feed Group items tagged

krzysieknowak

Deep Neural Networks are Easily Fooled: High Confidence Predictions for Unrecognizable ... - 4 views

  •  
    Other possible study: get a textbook example of an image of a pen, evolve it just enough so NN can't recognize it anymore, while minimizing the distance between the original and evolved images.
    EDIT: Its been done already: http://cs.nyu.edu/~zaremba/docs/understanding.pdf
  •  
    Of course, you can't really use them to extrapolate. The unknown unknown is always the trickiest :P
    They should just make another class "random bullshit", really and dump all of this stuff in there.
    I think there's a potential paper right there
Thijs Versloot

Real-Time Recognition and Profiling of Home Appliances through a Single Electricity Sensor - 3 views

  •  
    A personal interest of mine that I want to explore a bit more in the future. I just bought a ZigBee electricity monitor and I am wondering whether from the signal of the mains one could detect (reliably) the oven turning on, lights, etc. Probably requires Neural Network training. The idea would be to make a simple device which basically saves you money by telling you how much electricity you are wasting.

    Then again, its probably already done by Google...
  • ...3 more comments...
  •  
    nice project!
  •  
    For those interested, this is what/where I ordered.. http://openenergymonitor.org/emon/
  •  
    Update two.. RF chip is faulty and tonight I have to solder a new chip into place.. That's open-source hardware for you!
  •  
    haha, yep, that's it... but we can do better than that right! :)
Thijs Versloot

Thinking wind turbines - 0 views

  •  
    Siemens is using neural networks to improve operation of wind turbines, reducing maintenaince needs and improving output by one precent. It seems even that Siemens has quite a large neural network study group, probably linked to german universities, with various examples in practice (see websie)
Guido de Croon

Convolutional networks start to rule the world! - 2 views

  •  
    Recently, many competitions in the computer vision domain have been won by huge convolutional networks. In the image net competition, the convolutional network approach halves the error from ~30% to ~15%! Key changes that make this happen: weight-sharing to reduce the search space, and training with a massive GPU approach. (See also the work at IDSIA: http://www.idsia.ch/~juergen/vision.html)

    This should please Francisco :)
  • ...1 more comment...
  •  
    where is Francisco when one needs him ...
  •  
    ...mmmmm... they use 60 million parameters and 650,000 neurons on a task that one can somehow consider easier than (say) predicting a financial crisis ... still they get 15% of errors .... reminds me of a comic we saw once ... cat http://www.sarjis.info/stripit/abstruse-goose/496/the_singularity_is_way_over_there.png
  •  
    I think the ultimate solution is still to put a human brain in a jar and use it for pattern recognition.
    Maybe we should get a stagiaire for this..?
Christos Ampatzis

BBC NEWS | Health | A step closer to reading the mind - 3 views

  •  
    memory cloning
  •  
    "It would be very easy not to co-operate, and then it wouldn't work", that's still the important part. I'm sure Dario LOVES this paper. Would be nice to have a coffee with him right now...
ESA ACT

http://www.ai-junkie.com/ann/evolved/nnt2.html - 0 views

  •  
    Explanation of artificial neural networks (explains very nicely the basics of the programming behind Christos' PhD thesis)
ESA ACT

Nissan BR23C robot car: Japanese bee craziness - Crave at CNET.co.uk - 0 views

  •  
    An short article about something called biomimetics. To all I understand, it is not.
1 - 11 of 11
Showing 20 items per page